This is not a list of results. It’s a list of findings, and the difference matters. A result is something you have to take my word for. A finding is something you can check at your own desk.
In a small market, what identifies a company is not its performance: it’s its structure. Its industry, its size, how many units it has. That’s why the findings on this page are published as proportions and never as amounts.
The figures exist. They are in each client’s case file, with their source, their account and their period. They are the client’s, not ours, and they don’t leave that file.
Specialized services company
Once the real contribution margin was built, it turned out the company was 8.5% below its break-even point, not the distance everyone inside assumed.
It was not a restructuring. It was too little in sales.
What would flip it: a material item now classified as a fixed cost turning out to be variable. The line between fixed and variable is a judgment, not a reading, and it is stated in the case file.
The days of operation the company had ahead of it varied by almost eight times depending on a single decision: whether or not it could borrow against what its customers owed it.
No one had had that conversation.
What would flip it: the bank’s answer. If that collateral isn’t possible, there is no range: there is a single number, and it’s the low one.
Financial services company
There were no financial statements to read. All eight full sets came out of it, and all eight balanced to zero.
Balancing is not the same as being right. The balance check measures whether the set of statements is consistent with itself, not whether the books are correct. Those are two different questions, and this is the first. The second is answered later, and the answer is not always the one you expect.
What would flip it: one of the mapping reclassifications that had to be made turning out to be in the wrong block. That wouldn’t move the balance check, which is mechanical. It would move the reading.
Agribusiness company
It held more than five times as many months of raw material as of finished product, and the finished-product cushion covered half the time it takes to produce it.
Nor was there any seasonality to follow: the inventory was three times more stable than the demand it served, and the same calendar month was worth anywhere from half to double from one year to the next.
The fix was neither buying nor selling. It was converting what was already inside.
What would flip it: the real import replenishment lead time, which is not measured today. The target level was calculated under three sets of assumptions, from the most aggressive to the most conservative, and under all three there is excess inventory. What changes with the assumption is the size, not the direction.
The stretch the customer saw became almost twice as fast. The stretch no one measured doubled. The full cycle, which is what the customer actually waits for, barely moved: less than ten percent.
And the same decision as in the previous finding fixes it, without buying anything.
What would flip it: the line between the two stretches being drawn in the wrong place, meaning that part of what was counted as production is actually the customer waiting.
The two methods gave values that differed by more than half. The wrong one was the discounted cash flow: three quarters of its result hung on a terminal value built on a figure that was later corrected.
This doesn’t say discounted cash flow is always wrong. It says this one was, for this specific reason.
What would flip it: the discount rate is the owner’s assumption, not a measurement. With the rate that would fit this geography and this customer concentration, the result changes materially. That is still open.
The two strongest findings of this practice are not here, and they won’t be. They are recording errors in identifiable companies. Telling them with figures, or even by industry, would expose those companies.
All that can be said about them is their shape:
There is a kind of error a trial balance cannot reveal, because the entry balances perfectly. It balances, and it still records money that never came in.
If that sounds like something that could be happening in your company, that is exactly the conversation worth having.
Everything on this page went through three gates before it went out. None of them is optional, and none depends on someone remembering to apply it.
If nothing would flip it, it isn’t a finding: it’s an opinion. That’s why every finding on this page has its own written down.
The first conversation is to understand, not to sell.